Talk:Inherently funny word

There seems to be some confusion over appropriate sourcing for this article. According to Wikipedia's policies, Sources must support the material clearly and directly, and Articles may not contain any new analysis or synthesis of published material that serves to reach or imply a conclusion not clearly stated by the sources themselves .

Lead section: There are problems with the source recently added to the second sentence in the lead – the source quotes Mark Nutter of Chicago's Friends of the Zoo troupe stating that 'Words with a "K" or a hard "C" sound seem to be funny', Steven Wright stating, 'The "rule of three" and "the rule of K"—those are true', and Jerry Seinfeld stating, 'There`s the "rule of three," the "rule of K" and the rest is management'.^[3] Using this source to reach any conclusions regarding 'a range of influential comedians' enhancing the humor of their routines is improper editorial synthesis, a form of original research – the source does not name any of these comedians as 'influential', nor does it state that the humor of their routines was 'enhanced'.

Also, a single 'study of the "k words are funny" issue' doesn't verify anything about k being 'often cited', let alone 'the mythology of actors and writers' or the 'consonant plosives' p, b, t, d, k, and g. None of this is directly mentioned in the source used.^[4]

It should also be noted that there is not an exception to citation requirements specific to article introductions per WP:LEADCITE. More importantly, statements that are not substantiated elsewhere in the article are not exempt from verifiability requirements simply because they appear in the introduction. Per WP:VERIFY, all material in articles must be attributable to reliable, published sources.

Giving examples: Regarding the statement 'we need examples to show that this concept is widely held', collecting disparate 'examples' of the supposed phenomenon of inherently funny words in order to state or imply that 'the concept is widely held' is the definition of editorial synthesis. In a similar vein, the {{citation needed}} tag that I added to the introduction has nothing to do with providing 'examples'; once again, the issue is finding a source that directly supports the material per WP:VERIFY by stating that this 'common trope' is used in 'many fictional works'.

The sources added to provide such examples are also essentially all primary sources for the statements made. Relying on such primary sources can be problematic in that original research is required to analyze and interpret them. On the other hand, simply listing isolated excerpts is not the function of an encyclopedia. A Wikipedia article should not be a complete exposition of all possible details, but a summary of accepted knowledge regarding its subject. Verifiable and sourced statements should be treated with appropriate weight. It is precisely for the purposes of establishing 'accepted knowledge' and 'appropriate weight' that reliable, secondary sources are needed that directly address the topic and provide context. —Sangdeboeuf (talk) 01:41, 30 December 2016 (UTC)

Update: I trimmed the article down to remove material based solely on original research and primary sources, as well as material tagged in December 2016 as based on user-generated sources. —Sangdeboeuf (talk) 11:23, 23 January 2017 (UTC)

For future editors, note that this article was formerly substantially longer, and was trimmed down repeatedly (here, here, and here) by Sangdeboeuf sometimes with disagreement from other editors (here and here). I, for one, am not sure that some valuable contributions may have been lost in the process. Consider looking there for additional information and references. Sondra.kinsey (talk) 15:41, 17 June 2017 (UTC)

Several published how-to guides claim that k words are inherently funny.^[1][2][3][4] However, I don't think these sources are reliable here, since the authors all have a commercial interest in promoting the idea. Wikipedia is not a how-to guide, and it's strange that the only book sources covering the phenomenon fall into that category; if k words being funny were a genuine phenomenon, I'd assume that more general-interest sources would have picked up on it. Also, there's no way to verify that the idea of k words being funny isn't just hearsay reinforced through repetition. A different how-to-guide author calls the idea a 'myth', saying, 'The fact is that in the hands – or mouths – of a skilled comedian almost any word can be made funny'.^[5]

The author of a book about the show Seinfeld notes in passing that, 'That plosive consonant K sound is known to be among the English language's funniest phonemes',^[6] but once again, if it's so widely known, it's strange that there are so few reliable sources covering it. The idea that k words are inherently funny might simply be lifted directly from a previous version of this article, as in the 'Grammar Girl' podcast titled 'Words that sound funny'. In that case, what was initially original research by an unnamed Wikipedian becomes accepted 'knowledge' without any substantiation other than simple repetition. —Sangdeboeuf (talk) 12:17, 23 January 2017 (UTC)

It's certainly true that wikipedia articles can invisibility circle around online to become sources for themselves, but you've created that problem here by removing legitimate sources that get around that issue - e.g., Neil Simon's "The Sunshine Boys" of 1972 which mentions the k-words-are-funny concept way before the internet.

Your understanding of acceptable sources as reflected above is, IMHO, ridiculously narrow and academic, and will kill off this article. That would be too bad, but many years on wikipedia has taught me that beating my head against WP:I don't like it is more painful that it's worth.

Have fun. - DavidWBrooks (talk) 17:34, 23 January 2017 (UTC)

I added material on the part about k words from The Sunshine Boys, with references to reliable, secondary sources. The earlier problem with the way the play was mentioned in the article was that there were no such secondary sources cited to put the information in any context. As mentioned above, secondary sources should form the basis of Wikipedia articles. —Sangdeboeuf (talk) 20:36, 23 January 2017 (UTC)

As for killing off the article, that's fine with me. A lack of information is better than misleading or false information. Readers must be able to check that any of the information within Wikipedia articles is not just made up. —Sangdeboeuf (talk) 06:14, 24 January 2017 (UTC)

How though? Howrudoingmahg (talk) 12:55, 15 February 2025 (UTC)

The reliable sources that describe the science of "funny words" here^[1][2][3] refer to a single paper by Chris Westbury et al. which describes "perceived humor" (not inherent humor) as "a quantifiable function of how far NWs [non-words] are from being words",^[4] so the entire concept of "inherently funny words" seems based on a misinterpretation of Westbury's conclusions. Even with these few couple of articles, there simply isn't significant coverage in enough reliable, secondary sources to justify an encyclopedia article about the concept, in my opinion.

The part quoting from The Sunshine Boys shows that "vaudeville lore" held certain words to be funny, but it doesn't prove that inherently funny words exist.^[5] In a hundred years, when no one knows who or what Alka-Seltzer or Casey Stengel are, will anyone find these words funny? I propose merging the research portion into Humor research and the part about vaudeville into Humor (or possibly Comedy), with a redirect to either page, leaning towards the latter. —Sangdeboeuf (talk) 02:22, 15 February 2017 (UTC) (updated 05:54, 18 June 2017 (UTC))

I've posted notices at WikiProject Comedy and WikiProject Philosophy about this discussion. —Sangdeboeuf (talk) 02:45, 15 February 2017 (UTC)

I'm opposed, for what it's worth. I think that the topic is of interest and would be entirely lost after a merge. We could use more sources but I'm not convinced that there are no others. - CRGreathouse (t | c) 09:46, 19 March 2017 (UTC)

If you can demonstrate notability with significant coverage in reliable, secondary sources that contain analysis, evaluation, interpretation, or synthesis of information concerning the topic, then by all means add the sources to the article. —Sangdeboeuf (talk) 22:02, 19 March 2017 (UTC)

I just came to this page because Jimbo recently said again that this was his favourite article. We also see multiple accolades in external media coverage. The article has been nominated for deletion four times but the result each time was to keep it. The most recent nomination by Sangdeboeuf was likewise defeated with no-one supporting the proposal. But Sangedeboef still hasn't got the message and continues to try to destroy the article. There's no consensus for this merge proposal or any of his other updates and so I'm reverting them all. Andrew D. (talk) 16:33, 1 May 2017 (UTC)

See above comments about significant coverage of the topic (not the article) in reliable sources. "Accolades" in external media are irrelevant to whether an article meets Wikpedia's policies and guidelines. Likewise how Jimmy Wales happens to feel about a particular article doesn't mean anything about the fate of said article. —Sangdeboeuf (talk) 07:29, 11 May 2017 (UTC)

For the record, my nomination of this page at AfD was not "defeated" – I withdrew it for procedural reasons, such as considering the merger of some of the material into a different article first. Hence this discussion. —Sangdeboeuf (talk) 07:59, 11 May 2017 (UTC)

Oppose This is a recurring theme mentioned in by many sources, although Sangdeboeuf has deleted many of them. I recognize that only minimal research has been done, but this is a well-bounded topic, very appropriate for an article. What currently exists in this article currently achieves notability in my eyes. Sondra.kinsey (talk) 15:33, 17 June 2017 (UTC)

Per WP:NOTNEWS, short-lived coverage in the news doesn't automatically qualify something for an encyclopedia article. And to establish notability, WP:GNG requires more than trivial mentions of a topic. I haven't found any reliable, independent, secondary sources that address the subject in detail beyond the few about the one study by Westbury et al. already cited in the article. If there are more, please add them. —Sangdeboeuf (talk) 12:38, 19 June 2017 (UTC)

Oppose merge; since the discussion was last active, there have been several reverenced additions to a wider range of sources, suggesting that the topic does not reflect just a short-lived news story. Any objections to a close for a discussion with arguments which were valid at the time, but are no longer so? Klbrain (talk) 21:14, 30 July 2018 (UTC)

I'm collapsing the {{press}} template at the top of this talk page – I don't know whether or not all the message boxes recently added are intended to convince people that the article is really important by overwhelming them visually, but reducing the clutter would help facilitate the purpose of this page, which is the discussion of sources and policy. —Sangdeboeuf (talk) 09:03, 18 June 2017 (UTC)

Do you happen to know why this article has become very popular to read over the last two months? Take a look at the page view statistics. But the most recent press article listed above is 9 months old, and I didn't find anything new in a quick web news search. —BarrelProof (talk) 22:02, 18 June 2017 (UTC)

This continues to grow:

116 average views per day in the prior 3-month period (2017-01-26 – 2017-04-25)

528 average views per day during the next month (2017-04-26 – 2017-05-25)

1,488 after that until two days ago (2017-05-26 – 2017-06-18)

3,959 average views per day in the last two days (2017-06-18 – 2017-06-20)

—BarrelProof (talk) 03:33, 22 June 2017 (UTC)

There appears to be an unfortunate error in at least some of the secondary sources that discuss the article by Westbury, et al. Unfortunately, I don't have access to the full text of that article (and the Washington Post is also paywalled for me at the moment). I do see the article in The Guardian, which says "Well, what they've found is a strong inverse correlation between funniness and a property called entropy. This is a way of expressing how usual the letters in the NWs [nonsense words] are – so the less commonly they're used in English, the lower the total entropy of the NW. To put it another way, the less 'wordy' these NWs are, the more they strike us as humorous. ... The more it violates our expectations – again, the less 'wordy' it is – the more funny we find it. The authors also ran a quick analysis of some of the NWs used in the Dr Seuss books – like rumbus, skritz and yuzz-a-ma-tuzz – and found that they score low on entropy, making them particularly funny."

That is rather strange, as anyone who has studied Shannon entropy at all will tell you that when something is less probable, its entropy (more precisely, the amount of "information" that it conveys, also called the "surprisal") is higher, not lower. I therefore strongly suspect that what Westbury et al. said is being misrepresented in that press article. Properly, there would be an asserted inverse relationship between funniness and probability, but not between funniness and entropy.

There's also a bit of confusion evident here between entropy and surprisal. What is being discussed is the amount of surprise upon encountering a particular word, expressed mathematically as I(x) for a particular word x, not the entropy (denoted typically as H), which is the expected value of I(x) when considering all possible values of x.

—BarrelProof (talk) 20:38, 19 June 2017 (UTC)

It may be a mistake with regard to the mathematics of entropy, but it's consistent with what Westbury says in the original press release – that non-words with unlikely letter combinations, such as Dr. Seuss's creations, have lower entropy. In any event, we must stick to what the sources actually say. —Sangdeboeuf (talk) 21:11, 19 June 2017 (UTC)

For that matter, if you want to create a NW that is really highly improbable, instead of things that "make sense" in a way that a child could read properly like the Dr Seuss examples of "rumbus, skritz and yuzz-a-ma-tuzz", you would pick combinations of letters that really violate expectations, like "qpgzjx" or "wyjkbq", and those are not funny at all. They are strange, but they are not funny. —BarrelProof (talk) 21:19, 19 June 2017 (UTC)

Oh, I see that this last aspect has been considered – per the Smithsonian article: "As it turns out, there's a kind of "Goldilocks Zone" of nonsense words: A word like "anotain" got fewer laughs because it looks more like a real word, while "pranomp" got more because it looks just silly enough." The notion of a "Goldilocks Zone" seems to mean that a word needs to seem improbable, but not too improbable (like "qpgzjx" or "wyjkbq", presumably). —BarrelProof (talk) 17:24, 20 June 2017 (UTC)

I also found a Reader's Digest article that mentions that the nonsense words were specifically selected to be pronounceable and to not violate typical English spelling rules, which would also rule out "qpgzjx" and "wyjkbq". —BarrelProof (talk) 19:49, 20 June 2017 (UTC)

Section "Rudeness and entropy" claims that "whong", "dongl", "shart", and "focky" are funny because they are nonsense words with a shape similar to that of sexual slang. But at least two of them are in fact sexual slang: dongle is a diminutive of dong which means penis (see Donglegate), and shart means wet flatulence. So how are they "harmless nonsense"? --Damian Yerrick (talk) 12:22, 18 July 2018 (UTC)

@Damian Yerrick There is nothing wrong with sexuality existing in Wikipedia Jacobacademy (talk) 14:51, 21 September 2024 (UTC)

This article was the subject of a Wiki Education Foundation-supported course assignment, between 24 August 2020 and 9 December 2020. Further details are available on the course page. Student editor(s): Meganmckenna1. Peer reviewers: Mustafakhaled6.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 00:32, 17 January 2022 (UTC)

The paragraph citing two tech blogs recounting a single conference paper at a machine learning conference is baffling at best and doesn't seem to contribute anything to the article. Of course if you train a model on clusters of words and survey results, it can produce basic statistics like "highest ranked words" or predict similar survey results given new words.

Does this paragraph contribute anything other than mentioning "AI algorithms"? 2601:1C2:1600:E22:D105:A453:E2C8:1C15 (talk) 02:07, 12 February 2025 (UTC)